From: C L Freeman (
Date: Thu Feb 07 2008 - 04:53:30 CST

Thanks for your help Axel.

I went back and checked through the file some more but still could not find any
problem atom positions so I played around a little. If I deleted to last 3000
atoms to reduce its size to 99,999 atoms it was fine to view in vmd. So I
tried reducing the file to only those 3000 atoms and again it claimed I had
problems with the bonds. But this section could happily be visualised from
before it was combined to make the whole system. Then I realised in this
earlier case the atom numbers would have been less than 100,000. The problem
stems from my formatting of the pdb file. In my problem file I have:

ATOM 100000 etc.

clearly vmd can read a 6 digit atom number but it expects this atom number to
start only 1 space after ATOM not 2. Persumably it has been coded with
dividing lines into specfic numbers of characters as each term? So when I
changed it to:

ATOM 100000 etc.

I could visualise everything fine.

Thanks again for your help

Quoting Axel Kohlmeyer <>:

> On Wed, 6 Feb 2008, C L Freeman wrote:
> CF> Hello,
> colin,
> CF> I'm currently trying to use vmd to visualise a system composed of
> CF> some 28,000 calcite atoms with a protein/water mix of some 75,000
> calcite "atoms"?
> CF> atoms so a total system of 103,000 atoms. vmd returns an error when
> CF> I try and load the pdb file to view this structure I get an error
> CF> claiming some of my atoms have exceeded the maximum number of bonds.
> the way to get around this would be to provide a file with the
> bonding information (or recompile VMD to support more bonds).
> the warning about excessive bonds usually means that some atoms
> are close to overlapping. has your system been run as an MD or
> geometry optimization already?
> CF> Now I know this is not the case as I can load the two structures
> CF> indepently with no problems as none of the atoms have more than 4
> CF> bonds. If I try to load one structure and then the other I then get
> CF> the error:
> CF>
> CF> "Incorrect number of atoms (the number of atoms in the file)"
> you have to understand the logic of "molecules" and "frames" in VMD.
> it is not always obvious.
> a "molecule" in VMD is one structure, usually defined by a topology
> file like .psf or .parmtop (which can actually consist of many
> molecules). a "frame" is a set of coordinates for this "molecule".
> for loading .pdb (or .xyz or similar) files, VMD considers them
> to contain both and it will try to guess the missing structure
> information (e.g. bonds) based on internal heuristics.
> if you load a second file into an existing "molecule" VMD assumes
> that you want to load another frame (i.e. only coordinates) into
> an existing molecules and for that to work the second file has
> to contain the exact same number of coordinates. the test for this
> is an internal safety check. so the only way to have a merged
> structure is combinding the two .pdb files and loading them
> as one. so you should check of unusually close atoms and generally
> check your combined file. if you still are convinced that the
> file is 100% correct, please upload it to the VMD public biocore
> project (instructions are on the VMD homepage) so that VMD
> developers can have a look at it.
> hope that helps.
> cheers,
> axel.
> CF>
> CF> This suggests to me that there may be some issue with regard to the
> number of
> CF> atoms allowed? Is there a parameter I can change to account for this?
> CF>
> CF> Cheers
> CF> Colin
> CF>
> --
> =======================================================================
> Axel Kohlmeyer
> Center for Molecular Modeling -- University of Pennsylvania
> Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
> tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425
> =======================================================================
> If you make something idiot-proof, the universe creates a better idiot.